The goal of this project is to determine the role of serine, cysteine and metallo-proteases in the development and progression of epithelial cancers and to test the effects of anti-protease therapy in inhibiting these processes. In the first aim protease involvement in primary epithelial tumor growth, invasion and metastasis will be investigated using human prostate and colon cancer tumor cells in vitro and in vivo in xenograft models in immunocompromised mice. We will determine the serine, cysteine and metallo-proteases expressed in human tumors of prostate and colon cell lines derived from them with Core B. In the second aim, we will detrmin the effect in vivo of macromolecular protease inhibitors developed in Project 1 and Core C and chemical protease inhibitors deved in Project 1. These agents will either be administered by injection or expressed as transgenic recombinant proteins. Administration of exogenous agents will be guided by pharmakokinetic studies carried out by Core D. Protease and/or protease inhibitors will be modulated in transgenic mice by genetic complementation. We will use transgenic mice developed by Core C and Project 1 that are null or overexpress specific proteases or protease inhibitors and determine whether and when these proteases are involved in the multistage progression of malignancy. Tumor lines will be transplanted into immunocompromised mice that overexpress proteaseinhibitor transgenes on constitutive or temporally regulated promoters. In the third aim we will determine which proteases are expressed by endothelial cells responding to angiogenic stimuli in vivo and what effect protease inhibitors have on an in vivo model of growth factor-induced angiogenesis. We will then test these inhibitors for effects on inhibiting angiogenesis in xenograft models of human colon and prostate cancer. We believe that these studies will lead to novel chemotherapeutic agents with greater specificity and reduced toxicity than many currently used agents for these tumors.